Classification of microarray cancer data using ensemble approach

Abstract

An ensemble of classifiers is created by combining predictions of multiple component classifiers for improving prediction performance. In this paper, we conduct experimental comparison of J48, NB, IBK on nine microarray cancer datasets and also analyze their performance with Bagging, Boosting and Stack Generalization. The experimental results show that all ensemble methods outperform the individual classification methods. We then present a method, referred to as SD-EnClass, for combining classifiers from different classification families into an ensemble, based on a simple estimation of each classifier's class performance. The experimental results show that the proposed model improves classification accuracy, in comparison to simply selecting the best classifier in the combination. In the second stage, we combine the results of our proposed method with the results of Boosting, Bagging and Stacking using the combining method proposed, to obtain results which are significantly better than using Boosting, Bagging or Stacking alone.

An Ensemble of Filters and Wrappers for Microarray Data Classification

“…Then, the classifier with the highest class performance for a certain class out of the base classifiers becomes the expert of that class. The class specific performance of a classifier is calculated as: [11] Class specific accuracy= (Total no. of correctly predicted instances for a class)/ (total no.…”

“…Canedo et al [10] described a new framework for feature selection consisting of an ensemble of filters and classifiers. And also Nagi et al [11] combined the results of Boosting, Bagging and Stacking to obtain results which are significantly better than using Boosting, Bagging or Stacking alone. Liu et al [12] proposed a new ensemble gene selection method in which each gene subset is obtained by the same gene selector with different starting point.…”

“…Finally, after having good subset, it is enough that, algorithm 4 is applied in order to classify instance. For each gene in a specific dataset do (4) Calculate the statistical scores by that special filter (5) Rank the scores from the highest to the lowest For First set of features for each dataset do (9) Obtain subset of feature as a second set of features for each dataset by that special filter (10) For each learner do (11) For each set of subset that has obtained by each filter do //==there are five sets of feature==// (12) For second set of features for each dataset do (13) Calculate the accuracy by that special learner (14) Calculate the average of accuracy on that special set (15) Rank the average of accuracy from the highest to the lowest (16) Select the first set of subset for that learner Algorithm 3. Apply Wrappers.…”

Random Subspace Aggregation for Cancer Prediction with Gene Expression Profiles

“…Bolón-Canedo et al provided a novel framework for feature selection by an ensemble of filters and classifiers [10]. Combining classifiers from different classification families into an ensemble based on the evaluation of performance of each classifier, Nagi and Bhattacharyya proposed an ensemble method named as SD-EnClass [11]. To ensure a high classification accuracy, Ghorai et al showed an ensemble of nonparallel plane proximal classifiers based on the genetic algorithm through simultaneous feature and model selection scheme [12].…”

Cancer microarray data feature selection using Multi-Objective Binary Particle Swarm Optimization algorithm

“…The embedded techniques (Wahid et al, 2011[32]) allow interaction of different class of learning algorithms. More recently, the ensemble model (Nagi and Bhattacharyya, 2013[25]) based on different sub sampling strategies, the learning algorithms run on a number of sub samples and the acquired features are united into a stable subset. However the feature selection techniques can be also categorized based on search strategies used such as forward selection, backward elimination, forward stepwise selection, backward stepwise selection and random mutation (Mladeni, 2006[24]).…”